Week 7: Image-based Spam E-mail

今天的第一堂課, 黃老師首先在白板上寫了一個演算法, 要同學試著將程式寫出來。
在第二堂課, 幾乎所有同學都完成了程式的撰寫, 我們便開始向同學解說為什麼要同學寫這個程式。其實, 這個程式是下列論文在判斷一張影像是否屬於 Image Spam 的判斷特徵之一。

Image Analysis for Efficient Categorization of Image-based Spam E-mail

我們大致說明了 spam e-mail, 尤其是 image-based spam e-mail, 目前的發展, 然後請同學試著閱讀下面這段文章, 培養同學閱讀科技論文的能力。

Color saturation features:
As defined by Frankel et al., color saturation is quantified as the fraction of the total number of pixels in the image for which the difference max(R,G,B) – min(R,G,B) is greater than some threshold T (set to 50 by Frankel et al.; Hu and Bagga; and in this work). We evaluate this fraction for both text and non-text parts of the image separately, leading to two color saturation features. When compared with images of natural scenes, we expect the spam images to be generally more saturated due to the presence of synthetic graphics. However, when compared with generic computer-generated graphics images, we expect the spam images to be less saturated due to the presence of natural elements.